O'Halloran, Isabelle Rose (2025) Language Learning in Adults: A Role for Offline Memory Consolidation and Prior Knowledge. Doctoral thesis, York St John University.
Preview |
Text (Doctoral thesis)
Language Learning in Adults - A Role for Offline Memory Consolidation and Prior Knowledge.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. | Preview |
Abstract
Sleep plays an active role in the long-term consolidation of new words (Schreiner & Rasch, 2015), but its role in grammar learning is unclear, with some studies reporting benefits (Batterink et al., 2014) and others not (Mirković & Gaskell, 2016). However, previous research has not considered how prior knowledge may influence offline consolidation in grammar learning. For instance, the Complementary Learning Systems model (CLS; McClelland, 2013) posits that existing knowledge should facilitate neocortical integration of related information at encoding. Conversely, for new unrelated knowledge, consolidation is predicted to be slower, requiring integration to happen offline. Across four experiments, we examined the role of prior knowledge in grammar learning and investigated to what extent it influences initial learning and consolidation. Participants were trained and tested on an artificial language consisting of novel two-word phrases (e.g., tib viffeem) referring to a familiar picture (e.g., waitress). Systematic phonological properties (tib, eem) cued grammatical categories, and varied in their relation to existing knowledge, while arbitrary properties (i.e., new words) were reflected in the form-meaning mappings (viff-waitress). In Experiment 1, a recall task showed evidence of a role for sleep in the consolidation of the arbitrary mappings but no benefit for the consolidation of the systematic mappings, even for those mappings that were unrelated to existing knowledge. Consistent with CLS, participants showed better immediate recall for systematic mappings related to existing knowledge compared to those that were unrelated. This advantage at initial recall was replicated in Experiments 2 and 3 but took longer to emerge when training involved a more complex artificial language (Experiment 4). Moreover, a prior knowledge advantage at initial learning was not observed in online measures (Experiments 1 & 4) and did not extend to generalisation (Experiments 1- 4). The findings are discussed in relation to models of memory consolidation.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Status: | Published |
| Subjects: | B Philosophy. Psychology. Religion > BF Psychology |
| School/Department: | School of Education, Language and Psychology |
| URI: | https://ray.yorksj.ac.uk/id/eprint/15390 |
University Staff: Request a correction | RaY Editors: Update this record
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)
